Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 5770
    Prev
    Next
    INDEX
    Explanations

    instances of user interface elements related to navigation and actions in a programming context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    next or start

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ficult
    -0.37
     inf
    -0.36
    Datagram
    -0.34
     Hutchinson
    -0.33
    Rema
    -0.30
     dobr
    -0.30
    Darryl
    -0.30
     Cien
    -0.30
    Remain
    -0.29
    깝
    -0.29
    POSITIVE LOGITS
     <>",
    0.59
    tonode
    0.57
    RTSC
    0.52
     betweenstory
    0.52
    ollectionView
    0.52
     Vikipedi
    0.51
     NavController
    0.50
    igurumi
    0.50
     HasFactory
    0.50
    EDEFAULT
    0.50
    Activations Density 0.004%

    No Known Activations